Cascade type encoding process with error detection and pack numbering for hierarchic arq systems
Abstract
A retransmission method is disclosed wherein retransmissions over the sensitive link of the connection only involves packets that were actually disturbed in the sensitive link. At the same time, required storage capacity in the relay stations connecting the sensitive link to the nonsensitive link is reduced. Errors occurring in the sensitive part of the connection will cause retransmission over nonsensitive links of the connection as well, but this is considered to be of less importance since the nonsensitive links have a high throughput and/or low costs.

Term
Term ended
Expired 19 December 2016, 9.8 years ago.
- Priority
- Filed
- Granted
- Expired
- Today
11 claims: 3 independent, 8 dependent
- 1Zastrzeżenia patentowe 1. Sposób minimalizowania retransmisji pakietów informacji między dwiema stacjami końcowymi, w którym rozdziela się informację na numerowane pakiety i koduje się z detekcją błędu te pakiety z wykorzystaniem pierwszego kodu w pierwszej stacji końcowej, znamienny tym, że w kolejnych etapach nadaje się pakiety do stacji przekaźnikowej z wykorzystaniem łącza wrażliwego, gromadzi się pakiety w pamięci stacji przekaźnikowej, lokalnie koduje się, z detekcją błędu i numeruje odebrane pakiety, nadaje się pakiety do drugiej stacji końcowej z wykorzystaniem łącza niewrażliwego, dekoduje się lokalny kod detekcji błędu pakietu i potwierdza odbiór, jeżeli kod lokalny jest zdekodowany prawidłowo oraz dekoduje się pierwszy kod detekcji błędu i wysyła potwierdzenia do pierwszej stacji końcowej, jeżeli pierwszy kod detekcji błędu jest zdekodowany prawidłowo.
- 2Sposób według zastrz. 1, znamienny tym, że retransmituje się pakiet ze stacji przekaźnikowej do drugiej stacji końcowej w przypadku nieodebrania przez stację przekaźnikową potwierdzenia w określonym z góry okresie czasu po nadaniu pakietu.
- 3Sposób według zastrz. 1, znamienny tym, że retransmituje się pakiet z pierwszej stacji końcowej w przypadku nieodebrania przez stację końcową potwierdzenia w określonym z góry okresie czasu po nadaniu pakietu.
- 4Sposób według zastrz. 1, znamienny tym, że dekoduje się pierwszy kod detekcji błędu na stacji przekaźnikowej i potwierdza odbiór pakietu, jeżeli pierwszy kod detekcji błędu jest zdekodowany prawidłowo.
- 5Sposób według zastrz. 1, znamienny tym, że w potwierdzeniach wykorzystuje się numerację.
- 6Sposób minimalizowania retransmisji pakietów informacji między dwiema stacjami końcowymi, w którym rozdziela się informację na numerowane pakiety i koduje się z detekcją błędu te pakiety z wykorzystaniem pierwszego kodu w pierwszej stacji końcowej, znamienny tym, że w kolejnych etapach koduje się ten pakiet z detekcją błędu, z wykorzystaniem kodu lokalnego i numeruje się pakiet z wykorzystaniem numeru lokalnego, nadaje się zakodowany pakiet do stacji przekaźnikowej z wykorzystaniem łącza niewrażliwego, dekoduje się lokalny kod detekcji błędu odebranego pakietu i potwierdza się odbiór pakietu, jeżeli kod lokalny jest dekodowany prawidłowo usuwa się lokalny kod z detekcją błędu i lokalny numer ze zdekodowanego pakietu, nadaje się pakiet z usuniętym kodem i numerem do drugiej stacji końcowej przez łącze wrażliwe, dekoduje się pierwszy kod detekcji błędu i wysyła potwierdzenie odbioru, jeżeli pierwszy kod detekcji błędu jest zdekodowany prawidłowo.
- 7Sposób według zastrz. 6, znamienny tym, że retransmituje się pakiet z pierwszej stacji końcowej do stacji przekaźnikowej w przypadku nieodebrania przez pierwszą stację potwierdzenia w określonym z góry okresie czasu po nadaniu pakietu.
- 8Sposób według zastrz. 6, znamienny tym, że w potwierdzeniach wykorzystuje się numerację.
- 9Sposób minimalizowania retransmisji pakietów informacji między dwiema stacjami końcowymi, w którym rozdziela się informację na numerowane pakiety i koduje się z detekcją błędu te pakiety z wykorzystaniem pierwszego kodu w pierwszej stacji końcowej, znamienny tym, że w kolejnych etapach nadaje się pakiety do drugiej stacji końcowej przez zbiór łączy połączonych przez zbiór pośrednich stacji przekaźnikowych, przy czym w każdej stacji przekaźnikowej wprowadza się do odebranego pakietu dodatkowy, inny kod detekcji błędu i numerację i w każdej stacji przekaźnikowej dekoduje się lokalny kod detekcji błędu i usuwa się z odebranego pakietu zdekodowany kod korekcji błędu, a do stacji przekaźnikowej, która ostatnio nadawała pakiet, nadaje się potwierdzenie, jeżeli kod lokalny został zdekodowany poprawnie, tylko dla pakietów odebranych z wykorzystaniem łącza wrażliwego, 182 743 dekoduje się lokalne kody detekcji błędu pakietu i potwierdza się odbiór do odpowiedniej stacji przekaźnikowej, jeżeli kod lokalny jest zdekodowany prawidłowo oraz dekoduje się pierwszy kod detekcji błędu i wysyła się potwierdzenie do pierwszej stacji końcowej, jeżeli pierwszy kod detekcji błędu jest zdekodowany prawidłowo.
- 10Sposób według zastrz. 9, znamienny tym, że retransmituje się pakiet z jednej ze stacji przekaźnikowych do drugiej stacji końcowej w przypadku nieodebrania przez stację przekaźnikową potwierdzenia w określonym z góry okresie czasu po nadaniu pakietu.
- 11Sposób według zastrz. 9, znamienny tym, że retransmituje się pakiet z pierwszej stacji końcowej w przypadku nieodebrania przez pierwszą stację końcową potwierdzenia w określonym z góry okresie czasu po nadaniu pakietu. * * *
Independent claims11
57 paragraphs in 28 sections, as filed
The invention relates to a method for minimizing retransmission of information packets between two end stations. The invention relates to packet data transmission over a chain of cascaded links which use packet transmission for error correction, and in particular over cascaded links with different characteristics with regard to transmission costs per packet, transmission quality, transmission speed and / or telecommunications traffic volume. . Examples of such cascaded transmission links are portable computing devices (laptop, organizer, PDA) which are connected wirelessly (e.g. by radio link) to a mobile or portable telephone which in turn is connected via a radio link to the cellular network. Another example is wireless LAN where the computing device is connected wirelessly to a fixed site or satellite station attached to a wireless LAN (Ethernet) structure.
In high-speed transmission, packet-switched networks are used, in which data is carried in data packets, which, apart from information, carry source and destination addresses. When a packet is subject to interference, for example by collision with another packet or following an increased interference level, the packet from the source must be retransmitted again. The source must retransmit the packet until it successfully forwarded, as indicated by an acknowledgment from the destination that the packet was successfully received. Error detection code is added to the packet to be able to determine at the destination that the packet has been correctly received, either as Forward-Error-Correction FEC or Cyclic Redundancy Check (CRC). Redundancy Check).
There are several systems that provide automatic retransmission in the event that the destination does not acknowledge receipt of the packet, so-called Automatic Repeat Query (ARQ) systems. If the packet is not acknowledged within a certain period of time, the source automatically repeats the transmission. In the simplest case, the source expects an acknowledgment from the destination after transmitting each packet, and performs periodic repeating transmitting the same packet until receiving an acknowledgment. Sending the next package takes place only after receiving the confirmation of receipt. This is known as the stop-and-wait ARQ system.
More efficient methods perform transmitting the next packets even if the previous ones have not been acknowledged. Packets that have not been acknowledged are stored for retransmission and only deleted in memory after they have been acknowledged. In these methods, each packet is provided with a packet number, so that information about the correctly received packet numbers can be transmitted from the destination. Examples of ARQ methods that use packet numbering and do not require waiting for an acknowledgment before transmitting the next packet are: selective ARQ and (also cumulative) go-back-N ARQ. These methods provide higher throughputs, especially for connections with a certain amount of delay.
182 743
Most often, teledation is not carried out by means of a homogeneous connection, but through a chain of cascading links that are connected by relay stations. Very often, individual links have different properties and one link may be more sensitive than the others in terms of cost (public wired and cellular networks compared to local, private networks), data transmission speed or transmission quality (which may depend on the traffic intensity). movement). An example of such a cascaded connection is a connection between a cellular base station and a portable computing device that uses a short-range wireless connection to a cellular telephone to access the cellular base station. In this case, the cellular link is the most sensitive link. Another example is a portable computing device that is wirelessly attached to the Local Area NetWork (LAN) through which it connects to the server. In this example, the sensitive link is the wireless link between the computer and the wired LAN. In these cascading connections, the cross-ARQ protocol (ARQ only between source and destination; relay stations only pass the information without validating it) is not attractive because in the event of an error, retransmission is requested throughout the chain, regardless of where the error occurred. In order to minimize retransmission over a sensitive link, it is sometimes advisable through this sensitive link to retransmit only those packets that have been disturbed on that sensitive link. This requires distributed ARQ systems, that is, a separate ARQ system on each link. To optimize throughput, extensive memory capabilities are now required in the relay stations to provide effective ARQ methods.
The method of minimizing retransmission of information packets between two end stations, in which information is split into numbered packets and error-detected encoded these packets using the first code at the first end station, is distinguished by the fact that packets are transmitted to the station in subsequent steps. relay using a sensitive link, packets are collected in the relay station's memory, locally encoded, with error detection and numbers received packets, transmits packets to the second end station on a non-sensitive link, decodes the local packet error detection code and acknowledges reception if the local code is correctly decoded, and decodes the first error detection code and sends acknowledgments to the first station final if the first error detection code is decoded correctly.
Preferably, the packet is retransmitted from the relay station to the second end station in the event that the relay station does not receive an acknowledgment within a predetermined period of time after transmitting the packet.
Preferably, the packet is retransmitted from the first end station in the event that the end station does not receive an acknowledgment within a predetermined period of time after transmitting the packet.
The first error detection code is preferably decoded at the relay station and acknowledges receipt of the packet if the first error detection code is decoded correctly. Numbering is preferably used in confirmations.
The method of minimizing retransmission of information packets between two end stations, in which information is split into numbered packets and error-detection encoded those packets using the first code at the first end station, is distinguished by the fact that in subsequent steps this packet is encoded with error detection, using the local code and numbering the packet using the local number, transmit the encoded packet to the relay station using the non-sensitive link, decode the local error detection code of the received packet and acknowledge receipt of the packet, if the local code is decoded correctly, remove the local error detection code and the local number from the decoded packet, transmit the packet with the removed code and number to the second end station over the sensitive link, decode the first error detection code and send an acknowledgment of receipt, if the first error detection code is decoded correctly.
182 743
Preferably, the packet is retransmitted from the first end station to the relay station in case the first station did not receive an acknowledgment within a predetermined period of time after transmission of the packet. Numbering is preferably used in confirmations.
A method of minimizing retransmission of information packets between two end stations, in which information is split into numbered packets and error-detected encoded these packets using a first code at the first end station, according to the invention, end station via a set of links connected by a set of intermediate relay stations, wherein at each relay station an additional packet is introduced into the received packet, a different error detection code and numbering, and at each relay station the local error detection code is decoded and the decoded error correction code is removed from the received packet, and transmits to the relay station that last transmitted the packet confirmation is given, if the local code has been decoded correctly, only for packets received over the sensitive link, local packet error detection codes are decoded and acknowledged to be received to the respective relay station if the local code is decoded correctly and the first error detection code is decoded and an acknowledgment is sent to the first end station if the first error detection code is decoded correctly.
Preferably, a packet is retransmitted from one of the relay stations to the other end station in the event that the relay station does not receive an acknowledgment within a predetermined period of time after transmitting the packet. The packet from the first end station is preferably retransmitted in the event that the end station does not receive an acknowledgment within a predetermined period of time after transmitting the packet.
The present invention proposes a method in which only packets that have actually been disturbed on that sensitive link are retransmitted on a sensitive link. At the same time, the usually high capacity required of the relay stations connecting this sensitive link with the non-sensitive link is reduced. Errors occurring in the sensitive part of a connection will also result in retransmission over the non-sensitive links of the connection, but this is considered to be less important as the non-sensitive links are characterized by high bandwidth and / or low cost.
An ARQ crossover protocol is used between two end stations connected via a sensitive link and at least one non-sensitive link. The end stations have sufficient processing power and storage capacity to implement this cross protocol. Non-sensitive links between end stations use their own "local" ARQ methods which cause the packet to provision, along with a cross code and number, as a new information packet to which they add their own local fault detection code and number. Thus, the local ARQ protocol encapsulates the ARQ cross protocol and as a result a cascade of error correction coding and numbering is obtained. The packet carried over a sensitive link contains only cross-code and numbering. On the other hand, the packet carried over the non-sensitive link is additionally provided with a code and local number. A relay station between the non-sensitive and sensitive links that receives the packet from the non-sensitive link validates the received packet using local error-detection coding. When the relay station determines that the packet has been successfully received, it acknowledges the local number. Then the relay station takes the local code and number and forwards the packet to the sensitive link. If the packet has not been correctly received at the relay station, it is neither acknowledged nor forwarded to the sensitive link. When the relay station receives a packet from the sensitive link, it does not validate the received packet, but simply adds the code and local number and then forwards the packet to the non-sensitive link. At the end station acting as the destination, the local error detection code is checked first. If the result of the check is positive, it is acknowledged to the relay station. A cross protocol is then performed to see if the packet also passed the sensitive link without error. If the packet was not received correctly, no acknowledgment takes place, and the station at the other end retransmits the packet as the source. After correct receipt of the package, the destination acknowledges its receipt by providing the source with the number of this pa
182 743 cross brace. According to one embodiment of the present invention, a method is provided for minimizing retransmission over a sensitive link, combining high capacity with low power supply requirements at the relay stations. It is a combination of ARQ, cross and distributed protocols.
The subject of the invention, in an exemplary embodiment, is shown in the drawing, in which fig. 1 shows the transmission of a packet over a local network, fig. 2 - transmission of a packet via a cascade connection between a laptop computer and a cellular system via a short-haul high-frequency link to a mobile phone, fig. 3 - ARQ cross protocol, Fig. 4 - ARQ distributed protocol, and Fig. 5 illustrates a cascade protocol with error detection, according to one embodiment of the present invention.
Transmission links rarely consist of homogeneous links. Usually, transmission connections consist of a chain of cascading links, where each individual link has its own parameters that determine performance (data transmission speed and link quality) and transmission costs.
In data transmission, the methods with packetization are usually used to transfer data. This data is collected in a packet which may have some overhead of general information such as source and destination address, delivery priority and sequence number. Moreover, some form of error correction coding, for example cyclic redundancy check or FEC (forward-error-correction), is implemented so that the receiver is able to recognize whether the packet has been correctly received. Packets can then be transmitted over the link, either synchronously or asynchronously. At the destination, the correctness of the packet is checked and either an ACK acknowledgment is sent together with the packet number, or a non-acknowledgment signal (NAK), in the case of correct or incorrect receipt of the packet, respectively. After creating a non-acknowledgment signal, the source may respond by retransmitting the invalid packet. In many systems, the non-acknowledgment signal is not used. Instead, the source waits for a programmed period of time, and if no valid signal is received during that time, the source retransmits the packet. This so-called Automatic Repeat Query (ARQ), or automatic repeat request, is more reliable than the transmission of non-acknowledged signals because when the acknowledgment signal is disturbed, the source repeats the packet transmission, while when the non-acknowledgment signal is disturbed, some information is may not reach its destination. Note that acknowledgment messages do not have to be returned separately, but can be embedded in the return data stream as so-called reverse completions.
There are several variations to the ARQ procedure. In the simplest way, the source transmits only one packet and then waits for that packet to be acknowledged. The same packet is retransmitted periodically until it is confirmed. This stop-and-wait ARQ operation is not very efficient, especially when there is a significant delay in the connection or in processing at the destination. The ARQ method with higher bandwidth relies on sending packets continuously, but storing transmitted packets until they are acknowledged. After confirmation of the package, it is removed from the list of saved packages. If there is no acknowledgment of the packets within a certain period of time, they are retransmitted.
The invention also proposes a method for using these ARQ protocols in a chain of cascaded links or connections. Two examples of these protocols are shown in Figures 1 and 2. Figure 1 shows a wireless local area network (LAN). The server is connected to the wired LAN network, while the portable computer, which is the terminal, is connected to the same wired LAN via a radio link. In this example, the radio link is the sensitive link because it has a low bandwidth (it has a lower transmission speed, lower quality and therefore a greater number of retransmissions). The server and laptop in this example are both endpoints. The site acts as a relay station between the radio link and the wired LAN and can host several notebook computers. In Fig. 2, a portable device such as a computer of the type
182 743 The "laptop", PDA, organizer, etc. is connected via a shortwave radio link to a cellular telephone, which is connected via a radio link to the cellular network. In this example, the cellular link is sensitive to bandwidth (data stream speed and signal quality) and airtime costs. The cellular base station (or proxy unit attached to the base station) and the data device act as end stations, and the cell phone acts as a relay station. Due to the reduction of throughput and / or costs, errors in either the wired LAN (a) or the shortwave radio link (b) should not require retransmission over sensitive links.
ARQ protocols provide error correction by retransmission in the communication systems described above. The ARQ forward method uses some cross protocol in which only the end stations are checked for packet checking as shown in Figure 3. In Figure 3 the dashed lines represent a sensitive link. Only the end stations need to provide the memory capacity needed for the cross protocol. A relay station is only designed to relay information from one link to another without doing anything to the packets. This ARQ protocol is not attractive in the above-mentioned applications as errors in the insensitive link will result in cross retransmission including also the sensitive links. This drawback can be overcome by using the distributed ARQ protocol shown in Fig. 4. In the distributed ARQ protocol, each of the sensitive and non-sensitive links has its own ARQ protocol. Retransmissions take place only on the link where the errors actually occurred.
As shown in Fig. 4, the relay station must now be able to implement both local protocols, including frame storage for retransmission. For a non-sensitive link, this is not such a problem. Since the throughput is high and the loop delay is low, the buffering requirements are reduced as higher throughput and lower acknowledgment delay require smaller buffering sizes. On the other hand, in the case of a sensitive link, more developed memory capabilities are needed. However, for a wired LAN system that supports a large number of handheld devices concurrently, and for a handheld telephone, the additional memory and end protocol requirements are disadvantageous. In Fig. 4, this difference in memory capacity for the two ARQ methods is represented by the difference in memory sizes shown. In fig. 3-5, the packet characteristic parameters are shown in capital letters I, D and N, where I denotes the information packet, D denotes data entered to provide error correction and N denotes the packet number. In Fig. 3, there was only one type of packet which included error-detection cross-coding De and a Ne number, the subscript E signifying that the error-detecting coding D and the N number are part of the cross-protocol. In fig. 4 there are two local protocols producing Dli, Nli and Dli, Nlz for the two local ARQs, the lower index L denotes that the error detection encoding D and the N number are part of the local protocol. In Figure 4, it was assumed that the packet length was the same on both links. This is not a prerequisite and link 1, for example, could very well carry Dli (II), Nd, Dli (I2) Nli .., DLi (In) NLi by breaking the link 2 packet into n packets for link 1.
According to one aspect of the present invention, the complexity of a relay station can be reduced by using the local ARQ method for the insensitive link only. This method requires only a small amount of memory. This can be achieved by surrounding the ARQ cross protocol with a local ARQ protocol as shown in Figure 5. In addition to the cross protocol between these two end stations, a local ARQ protocol is used on the non-sensitive link. Cross-protocol packets are treated as normal data in the case of the local ARQ protocol and given additional encoding and numbering error. If end station A wishes to transmit a certain I packet, it adds the error detection code De and the cross protocol number Ne to it, receiving De (I) Ne in the packet. Currently, the local ARQ protocol introduces a second layer with the error detection code D1 and the number N<sub>NS</sub> to create a Dl package (D.<sub>e</sub>(I) Ne) N<sub>NS</sub>. Upon receipt of such a packet by the relay station, the error detection code is first checked to see if the link
182 Local 743 did not cause any errors. If the packet was received correctly, then the local number N<sub>NS</sub> it is acknowledged to station A. The relay station then removes the local ARQ (i.e. D1 and Nl) overhead from the packet and the packet is forwarded to the sensitive link. If the packet was not received correctly, neither the acknowledgment local number Nl nor the sensitive packet is forwarded. Station A then retransmits the packet locally over the non-sensitive link until properly received and acknowledged by the relay station.
After the forwarded packet has been received by end station B, the packet is checked for correctness using the error detection code De. This packet could possibly be affected by errors on the sensitive link, otherwise the packet would not be forwarded. Then, if the packet was received correctly, the cross-protocol packet Ne number is forwarded to station A in an acknowledgment. If the packet was not received correctly, it must be retransmitted throughout the connection. Since the local link is a high-capacity non-sensitive link, this additional forward traffic is not a problem on the non-sensitive link.
The transmission of a packet in the opposite direction is described below. End station B takes over certain information packet I and adds the fault detection code De and the cross protocol number Ne to it, obtaining De (I) N<sub>e</sub>. Then the coded packet is sent over the sensitive link to the relay station. The relay station does not check the correctness of receiving the packet. The packet is captured and constitutes the second layer with the error detection code D1 and the number Nl, superimposed on the packet forming the packet Dl (De (I) Ne) Nl- After receiving the packet, relay station A first checks if the packet went correctly over the last local link , by checking the error detection Dl- If the error detection code is correct, the end station A acknowledges receiving the (local) packet to the relay station N<sub>NS</sub>.
If the packet number N1 is not acknowledged, the packet is retransmitted but only by the relay station. If the error detection code D1 is correct, end station A proceeds to the next ARQ layer and checks the cross protocol error correction code De. If the error detection code De is correct, it is clear that the packet went through the entire call correctly and its reception can be acknowledged by giving the Ne number to end station B. If the error detection code De is not correct, then apparently an error has occurred on the sensitive link and the ending Ne packet number is not acknowledged. As a result, the endstation retransmits the packet until it is acknowledged by endstation A.
In another embodiment of the present invention, the relay station checks the error detection code De in the packet received from end station B. Upon detecting the error, it discards the packet, thus reducing the traffic on the non-sensitive link. While the excess ballast for sending invalid packets over a high-speed link is not a big deal, error checking in the end-protocol could unnecessarily burden the relay station.
Although the relay station still needs some memory capacity and computing power to implement the local procedure, however, due to the high bandwidth of the local links, the buffering and processing requirements are much lower for the local ARQ than for the cross ARQ. Also, since the local ARQ is embedded in the pass-through ARQ, any errors not detected in the local routine will be caught by the pass-through routine. It is clear that this additional security should be used as little as possible to minimize additional retransmissions over the sensitive link. However, a non-sensitive link does not have to be completely error-free, which further simplifies the implementation of local ARQ. One possible extension to this method is the arrangement in which the cross packet is split into smaller subpackets, each of which is locally encoded and numbered, and which are then transmitted at high speed over the local link. At the relay station, the subpackets are collected and assembled into a single packet, which is then transmitted over the sensitive link. Yet another extension of the basic method is the layout where several cross packets are collected and the evil one
182 743 wives into a large packet, which is then locally encoded and numbered. This large packet is then transmitted over a local (non-sensitive) link. At the relay station, a correctly received composite packet is disassembled into primary cross packets, which are then forwarded to and transmitted over the sensitive link.
The encapsulation procedure may continue if there are more cascaded and sensitive links. Each time, a new envelope of ARQ information is formed around the former packet. With a new encapsulation, the entire packet (info + code + number) is considered a new information packet. In this way, a certain hierarchy of ARQ circuits is formed and around the packet is arranged in the form of a layer, or overhead of individual ARQ circuits, one on top of the other, and the processing station can determine where the error has occurred and to which (relay) station the packet acknowledgment can be forwarded.
182 743
<img file="PL182743B1_D0001.tif" />
CELLULAR BASE STATION
182 743
Fig. 3
ARO SKINNY
<img file="PL182743B1_D0002.tif" />
END STATION A
RELAY STATION
<img file="PL182743B1_D0003.tif" />
END STATION B
<img file="PL182743B1_D0004.tif" />
D: CODING WITH ERROR DETECTION
I: INFORMATION
N: PACKAGE NUMBER
NON-SENSITIVE LINK
SENSITIVE JOIN
PACKAGE MEMORY
182 743
Fig. 4
ARO LOCAL
ARO LOCAL
<img file="PL182743B1_D0005.tif" />
RELAY STATION
END STATION A
<img file="PL182743B1_D0006.tif" />
END STATION B
<img file="PL182743B1_D0007.tif" />
D: CODING WITH ERROR DETECTION <sup>INCREDIBLE</sup> / INFORMATION --------- LINK SENSITIVE
PACKAGE MEMORY
N: PACKAGE NUMBER
182 743
Fig. 5
<img file="PL182743B1_D0008.tif" />
END STATION A
RELAY STATION
END STATION B
<img file="PL182743B1_D0009.tif" />
D: CODING WITH ERROR DETECTION
I: INFORMATION
N: PACKAGE NUMBER
------------ NON-SENSITIVE LINK
---------- SENSITIVE JOIN
PACKAGE MEMORY
182 743
Fig. /
LAN
<img file="PL182743B1_D0010.tif" />
Publishing Department of the Polish Patent Office. Circulation of 60 copies
Price PLN 4.00.
Contents28
15 sheets
Sheet 1 Sheet 2 Sheet 3 Sheet 4 Sheet 5 Sheet 6 Sheet 7 Sheet 8 Sheet 9 Sheet 10 Sheet 11 Sheet 12 Sheet 13 Sheet 14 Sheet 15
17 members in 11 offices
Priority claims8
| Document | Office | Kind | Date |
|---|---|---|---|
| 58111195 | United States of America | A | |
| 58111195 | United States of America | A | |
| 9601705 | Sweden | W | |
| 9601705 | Sweden | W | |
| 95581111 | – | – | – |
| 96SE9601705 | – | – | – |
| US19950581111 | – | – | – |
| WO1996SE01705 | – | – | – |
Members17
| Document | Office | Kind | |
|---|---|---|---|
| WO9724829A1 | World Intellectual Property Organization (WIPO) | A1 | |
| AU1323697A | Australia | A | |
| US5699367A | United States of America | A | |
| EP0870378A1 | European Patent Office (EPO) | A1 | |
| PL327599A1 | Poland | A1 | |
| CN1212098A | China | A | |
| BR9612365A | Brazil | A | |
| KR19990076825A | Republic of Korea | A | |
| AU714480B2 | Australia | B2 | |
| JP2000502852A | Japan | A | |
| EE03366B1 | Estonia | B1 | |
| PL182743B1This record | Poland | B1 | |
| EP0870378B1 | European Patent Office (EPO) | B1 | |
| DE69632147D1 | Germany | D1 | |
| KR100431228B1 | Republic of Korea | B1 | |
| DE69632147T2 | Germany | T2 | |
| JP3677297B2 | Japan | B2 |
Numbers
- Publication, DOCDB
- 182743
- Publication, EPODOC
- PL182743B
- Application
- 96327599
- Application, DOCDB
- 32759996
- Application, EPODOC
- PL19960327599
Titles2
- English
- CASCADE TYPE ENCODING PROCESS WITH ERROR DETECTION AND PACK NUMBERING FOR HIERARCHIC ARQ SYSTEMS
- Polish
- Sposób minimalizowania retransmisji pakietów informacji między dwiema stacjami końcowymi
Classification
- CPC, 7
- H04L1/1809
- H04L12/28
- H04L1/16
- H04L1/188
- H04L2001/0092
- H04L9/40
- H04L1/18
- IPC, 5
- H04L1 00
- H04L1 16
- H04L1 18
- H04L29 06
- H04L29 08